synthgen.py

import os
from openai import OpenAI

# Ensure the OPENROUTER_API_KEY environment variable is set
api_key = "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"
if not api_key:
    raise ValueError("OPENROUTER_API_KEY environment variable not set.")

# Point the OpenAI client to the OpenRouter API
client = OpenAI(
    base_url="https://openrouter.ai/api/v1",
    api_key=api_key,
)

def generate_synthetic_text(prompt: str, model: str = "deepseek/deepseek-chat-v3-0324:free") -> str:
    """
    Generates synthetic text using an OpenRouter model.

    Args:
        prompt: The input prompt to guide the text generation.
        model: The model to use on OpenRouter (default: gpt-3.5-turbo).
               You can find model names on the OpenRouter website.

    Returns:
        The generated text string.
    """
    try:
        response = client.chat.completions.create(
            extra_headers={
                # "HTTP-Referer": "https://www.google.com", # Optional. Site URL for rankings on openrouter.ai.
                "X-Title": "SynthGen", # Optional. Site title for rankings on openrouter.ai.
            },
            model=model,
            messages=[
                {"role": "system", "content": "You are a helpful assistant generating synthetic data."},
                {"role": "user", "content": prompt},
            ],
        )
        if response.choices and response.choices[0].message.content:
            return response.choices[0].message.content.strip()
        else:
            return "Error: No content generated."
    except Exception as e:
        return f"Error during API call: {e}"

# --- Main Execution ---
if __name__ == "__main__":
    # TODO: Define the kind of text and number of samples needed
    num_samples = 5 # Example: generate 5 samples
    prompt_template = "Generate a short, positive product review for a fictional gadget." # Example prompt

    print(f"Generating {num_samples} synthetic text samples...")

    for i in range(num_samples):
        # You might want to vary the prompt slightly for each sample
        # For now, we use the same template
        generated_text = generate_synthetic_text(prompt_template)
        print(f"\n--- Sample {i+1} ---")
        print(generated_text)

    print("\nGeneration complete.")